Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

Reinforcement Schedules01:24

Reinforcement Schedules

Positive reinforcement is a powerful method for teaching new behaviors to both animals and humans. B.F. Skinner demonstrated this with his experiments using rats in a Skinner box. When a rat pressed a lever, it received a food pellet. This immediate reward encouraged the rat to repeat the behavior. This method, where a reward follows every instance of the behavior, is known as continuous reinforcement. It is highly effective for establishing new behaviors quickly.
Once a behavior is learned,...
Reinforcement01:23

Reinforcement

Positive and negative reinforcement are key concepts in operant conditioning, a learning process where the consequences of a behavior affect the likelihood of that behavior being repeated.
Positive reinforcement occurs when a behavior is followed by the presentation of a rewarding stimulus, increasing the frequency of that behavior. For example:
Observational Learning01:12

Observational Learning

Albert Bandura's observational learning, also known as imitation or modeling, occurs when a person observes and imitates another's behavior. It is a quicker process than operant conditioning. A well-known example is the Bobo doll study, where children who saw an adult acting aggressively towards the doll were more likely to act aggressively when left alone, compared to those who observed a nonaggressive adult. Many psychologists view observational learning as a form of latent learning because...
Statically Indeterminate Problem Solving01:16

Statically Indeterminate Problem Solving

Statically indeterminate problems are those where statics alone can not determine the internal forces or reactions. Consider a structure comprising two cylindrical rods made of steel and brass. These rods are joined at point B and restrained by rigid supports at points A and C. Now, the reactions at points A and C and the deflection at point B are to be determined. This rod structure is classified as statically indeterminate as the structure has more supports than are necessary for maintaining...
Collisions in Multiple Dimensions: Problem Solving01:06

Collisions in Multiple Dimensions: Problem Solving

In multiple dimensions, the conservation of momentum applies in each direction independently. Hence, to solve collisions in multiple dimensions, we should write down the momentum conservation in each direction separately. To help understand collisions in multiple dimensions, consider an example.
A small car of mass 1,200 kg traveling east at 60 km/h collides at an intersection with a truck of mass 3,000 kg traveling due north at 40 km/h. The two vehicles are locked together. What is the...
Associative Learning01:27

Associative Learning

Associative learning is a fundamental concept in behavioral psychology, wherein a connection is established between two stimuli or events, leading to a learned response. This process is critical in understanding how behaviors are acquired and modified. Conditioning, the mechanism through which associations are formed, can be divided into two main types: classical conditioning and operant conditioning, each elucidating different aspects of associative learning.
Classical conditioning, also known...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Automatic robot design for fault-tolerant robots.

Scientific reports·2025
Same author

Existence of reservoir with finite-dimensional output for universal reservoir computing.

Scientific reports·2024
Same author

Automatic Temperature Parameter Tuning for Reinforcement Learning Using Path Integral Policy Improvement.

IEEE transactions on neural networks and learning systems·2023
Same author

Nonessentiality of Reservoir's Fading Memory for Universality of Reservoir Computing.

IEEE transactions on neural networks and learning systems·2023
Same author

Simultaneous Optimization of Discrete and Continuous Parameters Defining a Robot Morphology and Controller.

IEEE transactions on neural networks and learning systems·2023
Same author

Step Climbing Control of Snake Robot with Prismatic Joints.

Sensors (Basel, Switzerland)·2022

Related Experiment Videos

Dynamic structured pruning and merging for real-time acceleration of reinforcement learning.

Takato Ishii1, Ryo Ariizumi2, Fumitoshi Matsuno3

  • 1Department of Mechanical Systems Engineering, Tokyo University of Agriculture and Technology, Tokyo, 183-0057, Japan.

Scientific Reports
|July 12, 2026
PubMed
Summary

This study introduces a new method for efficient deep reinforcement learning (DRL) using dynamic structured pruning and model merging. It significantly reduces computational costs and latency for training and deploying DRL agents.

Keywords:
Deep reinforcement learningDynamic structured pruningModel merge

Related Experiment Videos

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Deep Reinforcement Learning

Background:

  • Deep reinforcement learning (DRL) faces high computational costs, limiting practical applications.
  • Current neural network pruning methods in DRL often use unstructured pruning, leading to hardware incompatibility and no real-time acceleration.
  • DRL pruning is difficult due to training instability and performance degradation.

Purpose of the Study:

  • To develop a novel framework for efficient DRL by integrating dynamic structured pruning with model merging.
  • To overcome the limitations of unstructured pruning and training instability in DRL.
  • To enable hardware-compatible structured pruning for accelerated DRL training and deployment.

Main Methods:

  • Proposed a framework combining dynamic structured pruning with periodic merging of parallel network instances.
  • Utilized structured pruning compatible with general-purpose hardware.
  • Implemented model merging to counteract instability from structural changes during pruning.

Main Results:

  • Reduced cumulative training FLOPs by up to 72% on continuous control tasks.
  • Achieved performance competitive with dense baselines in most environments.
  • Demonstrated an average inference latency reduction of 16.2%.

Conclusions:

  • The proposed framework effectively accelerates both training and deployment of DRL agents.
  • Dynamic structured pruning combined with model merging addresses computational costs and training instability in DRL.
  • The method offers a practical solution for deploying DRL in real-time applications.